Privacy Leakage in Federated Learning in Radiology Reports: Comparative Evaluation of Tokenizer and Batch-Size Privacy Risks.

Santhosh Parampottupadam^1,2, Andrés Martínez Mora¹, Dimitrios Bounias^1,2, Sinem Sav³, Klaus Maier-Hein^1,2,4, Ralf Floca¹

¹Division of Medical Image Computing, German Cancer Research Center, Im Neuenheimer Feld 280, Heidelberg, Baden-Wurttemberg, 69120, Germany, 49 06221 420.

²Medical Faculty Heidelberg, Heidelberg University, Heidelberg, Germany.

³Department of Computer Engineering, Bilkent University, Ankara, Türkiye.

⁴Pattern Analysis and Learning Group, Department of Radiation Oncology, Heidelberg University Hospital, Heidelberg, Germany.

PubMed DOI

Abstract

BACKGROUND: Federated learning (FL) enables multi-institutional model training on clinical text without sharing raw data; however, gradient inversion methods can reconstruct sensitive information from shared model updates. The extent of such privacy leakage in FL applied to radiology reports, and the role of tokenizer design, remains unclear. OBJECTIVE: This study aimed to quantify gradient-based…

Create a free account to keep reading

Free members get 10 full research views every month across publications, clinical trials, FDA clearances, adverse events, and NIH grants. No credit card required.

Create Free Account Sign In

Want unlimited research access? See Pro plans

Data Accuracy Notice: Research intelligence on Health AI Central is aggregated from public sources (PubMed, ClinicalTrials.gov, FDA, NIH, CMS, and others) and refreshed nightly. Classifications and derived metrics are produced by automated methods described in our Methodology. We recommend verifying critical data points against the primary sources before making decisions.